OpenAI revealed it had flagged and later banned the account of an individual months before he carried out one of Canada’s deadliest school shootings. While the activity raised concerns, it did not meet the threshold for referral to the Royal Canadian Mounted Police at the time.

In a troubling revelation, ChatGPT developer OpenAI disclosed that it had previously reviewed the account of an individual who would later carry out one of Canada’s deadliest school shootings.

According to the company, the account—belonging to Jesse Van Rootselaar—was flagged in June 2025 through internal safety systems for potential links to violent intent. At the time, OpenAI evaluated whether the case should be escalated to the Royal Canadian Mounted Police (RCMP). However, the activity did not meet the company’s threshold for law enforcement referral, which requires evidence of a credible and imminent threat of serious harm.

The account was ultimately banned for violating platform policies.

Months later, tragedy struck. The 18-year-old carried out a devastating attack in a remote community in British Columbia, killing eight people before taking their own life. Authorities reported that the incident began at the suspect’s home, where two family members were killed, before continuing at a nearby school.

Following the attack, OpenAI said it proactively contacted the RCMP and shared information related to the individual’s use of ChatGPT, pledging continued cooperation with the ongoing investigation.

The motive behind the violence remains unclear. The small mountain town, home to fewer than 3,000 residents, has since been left grappling with grief after the loss of students and school staff in what has become Canada’s deadliest mass violence incident in years.

Source: https://www.theguardian.com/world/2026/feb/21/tumbler-ridge-shooter-chatgpt-openai